Using Programmable Graphene Channels as Weights in Spin-Diffusive Neuromorphic Computing

A graphene-based spin-diffusive neural network is presented in this paper that takes advantage of the locally tunable spin transport of graphene and the non-volatility of nanomagnets. By using electrostatically gated graphene as spintronic synapses, a weighted summation operation can be performed in the spin domain while the weights can be programmed using circuits in the charge domain. Four-component spin/charge circuit simulations coupled to magnetic dynamics are used to show the feasibility of the neuron-synapse functionality and quantify the analog weighting capability of the graphene under different spin-relaxation mechanisms. This spin-diffusive neural network using a graphene-based synapse design achieves total energy consumption of 0.55–0.97 fJ per cell $\cdot $ synapse and attains significantly better scalability compared to its digital counterparts, particularly as the number and bit accuracy of the synapses increases.

[1]  Weisheng Zhao,et al.  Proposal for a graphene-based all-spin logic gate , 2015 .

[2]  J. Meindl,et al.  Conductance Modeling for Graphene Nanoribbon (GNR) Interconnects , 2007, IEEE Electron Device Letters.

[3]  P. Ruden,et al.  Rashba-induced spin scattering at graphene edges , 2013 .

[4]  Jacques-Olivier Klein,et al.  Analytical Macrospin Modeling of the Stochastic Switching Time of Spin-Transfer Torque Devices , 2015, IEEE Transactions on Electron Devices.

[5]  Takashi Kimura,et al.  Giant spin-accumulation signal and pure spin-current-induced reversible magnetization switching , 2008 .

[6]  Bernard Girau,et al.  Fault and Error Tolerance in Neural Networks: A Review , 2017, IEEE Access.

[7]  B. Wees,et al.  Eighty-Eight Percent Directional Guiding of Spin Currents with 90 μm Relaxation Length in Bilayer Graphene Using Carrier Drift. , 2016, Nano letters.

[8]  Don Monroe,et al.  Neuromorphic computing gets ready for the (really) big time , 2014, CACM.

[9]  Sachin S. Sapatnekar,et al.  CoMET: Composite-Input Magnetoelectric- Based Logic Technology , 2016, IEEE Journal on Exploratory Solid-State Computational Devices and Circuits.

[10]  An Chen,et al.  A Comprehensive Crossbar Array Model With Solutions for Line Resistance and Nonlinear Device Characteristics , 2013, IEEE Transactions on Electron Devices.

[11]  Jonathan Z. Sun.  Spin-current interaction with a monodomain magnetic body: A model study , 2000 .

[12]  Jacques-Olivier Klein,et al.  Spin-Transfer Torque Magnetic Memory as a Stochastic Memristive Synapse for Neuromorphic Systems , 2015, IEEE Transactions on Biomedical Circuits and Systems.

[13]  Xinjie Guo,et al.  Mixed Signal Neurocomputing Based on Floating-gate Memories , 2017 .

[14]  P. Kim,et al.  Energy band-gap engineering of graphene nanoribbons. , 2007, Physical review letters.

[15]  F. Guinea,et al.  Spin-orbit-mediated spin relaxation in graphene. , 2008, Physical review letters.

[16]  Takashi Taniguchi,et al.  Gate-tunable black phosphorus spin valve with nanosecond spin lifetimes , 2017, Nature Physics.

[17]  Xinjie Guo,et al.  Redesigning commercial floating-gate memory for analog computing applications , 2014, 2015 IEEE International Symposium on Circuits and Systems (ISCAS).

[18]  J. Appenzeller,et al.  Spin transfer torque in a graphene lateral spin valve assisted by an external magnetic field. , 2013, Nano letters.

[19]  Witold Pedrycz,et al.  Current-Mode Analog Adaptive Mechanism for Ultra-Low-Power Neural Networks , 2011, IEEE Transactions on Circuits and Systems II: Express Briefs.

[20]  Leon O. Chua,et al.  Cellular neural networks: applications , 1988 .

[21]  Claudia Felser,et al.  Mn3Ga, a compensated ferrimagnet with high Curie temperature and low magnetic moment for spin torque transfer applications , 2007 .

[22]  Behtash Behin-Aein,et al.  All-Spin Logic Device With Inbuilt Nonreciprocity , 2011, IEEE Transactions on Magnetics.

[23]  Catherine D. Schuman,et al.  A Survey of Neuromorphic Computing and Neural Networks in Hardware , 2017, ArXiv.

[24]  Dmitri E. Nikonov,et al.  Modeling and Design of Spintronic Integrated Circuits , 2012, IEEE Transactions on Circuits and Systems I: Regular Papers.

[25]  W. Rippard,et al.  Switching Distributions for Perpendicular Spin-Torque Devices Within the Macrospin Approximation , 2012, IEEE Transactions on Magnetics.

[26]  M. Kamalakar,et al.  Long distance spin communication in chemical vapour deposited graphene , 2015, Nature Communications.

[27]  Shigeyoshi Watanabe,et al.  A 0.67μm^2 self-aligned shallow trench isolation cell (SA-STI cell) for 3V-only 256Mbit NAND EEP-ROMs , 1994 .

[28]  R. Lippmann,et al.  An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[29]  Azad Naeemi,et al.  Evaluation of the Potential Performance of Graphene Nanoribbons as On-Chip Interconnects , 2013, Proceedings of the IEEE.

[30]  Andre K. Geim,et al.  The rise of graphene. , 2007, Nature materials.

[31]  R. S. Ross,et al.  Ultra-low resistance ohmic contacts in graphene field effect transistors , 2012 .

[32]  Azad Naeemi,et al.  A Proposal for Energy-Efficient Cellular Neural Network Based on Spintronic Devices , 2016, IEEE Transactions on Nanotechnology.

[33]  S. Koester,et al.  Contact-Induced Spin Relaxation in Graphene Nonlocal Spin Valves , 2016 .

[34]  Michael S. Fuhrer,et al.  Gate-tunable graphene spin valve , 2007 .

[35]  Sachin S. Sapatnekar,et al.  Spin-Based Computing: Device Concepts, Current Status, and a Case Study on a High-Performance Microprocessor , 2015, Proceedings of the IEEE.

[36]  Dan Hammerstrom.  DARPA neurocomputing , 2015, 2015 IEEE International Electron Devices Meeting (IEDM).

[37]  Dmitri E. Nikonov,et al.  Material Targets for Scaling All Spin Logic , 2012, ArXiv.

[38]  G. W. Burr,et al.  Experimental demonstration and tolerancing of a large-scale neural network (165,000 synapses), using phase-change memory as the synaptic weight element , 2015, 2014 IEEE International Electron Devices Meeting.

[39]  S. Datta,et al.  Proposal for an all-spin logic device with built-in memory. , 2010, Nature nanotechnology.

[40]  Spin relaxation in single-layer graphene with tunable mobility. , 2012, Nano letters.

[41]  Naresh R. Shanbhag,et al.  Shannon-inspired Statistical Computing to Enable Spintronics , 2017, ArXiv.

[42]  S. Koester,et al.  The effect of output-input isolation on the scaling and energy consumption of all-spin logic devices , 2015 .

[43]  K. Roy,et al.  Spin-Based Neuron Model With Domain-Wall Magnets as Synapse , 2012, IEEE Transactions on Nanotechnology.

[44]  Charbel Sakr,et al.  Analytical Guarantees on Numerical Precision of Deep Neural Networks , 2017, ICML.

[45]  R. Hecht-Nielsen,et al.  Theory of the Back Propagation Neural Network , 1989 .

[46]  Naveen Verma,et al.  Enabling system-level platform resilience through embedded data-driven inference capabilities in electronic devices , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[47]  J. Appenzeller,et al.  Improvement of spin transfer torque in asymmetric graphene devices. , 2014, ACS nano.

[48]  Howard C. Card,et al.  Stochastic Neural Computation I: Computational Elements , 2001, IEEE Trans. Computers.

[49]  Azad Naeemi,et al.  Non-Boolean Computing Benchmarking for Beyond-CMOS Devices Based on Cellular Neural Network , 2016, IEEE Journal on Exploratory Solid-State Computational Devices and Circuits.

[50]  Wei Han,et al.  Graphene spintronics. , 2014, Nature nanotechnology.

[51]  Wei Han,et al.  Tunneling spin injection into single layer graphene. , 2010, Physical review letters.

[52]  Weisheng Zhao,et al.  Gate-driven pure spin current in graphene , 2017, 2017 IEEE International Magnetics Conference (INTERMAG).

[53]  Masafumi Yamamoto,et al.  Giant tunneling magnetoresistance in epitaxial Co2MnSi/MgO/Co2MnSi magnetic tunnel junctions by half-metallicity of Co2MnSi and coherent tunneling , 2012 .

[54]  Byoungil Lee,et al.  Nanoelectronic programmable synapses based on phase change materials for brain-inspired computing. , 2012, Nano letters.

[55]  Dmitri E. Nikonov,et al.  Static and Clocked Spintronic Circuit Design and Simulation With Performance Analysis Relative to CMOS , 2014, IEEE Transactions on Circuits and Systems I: Regular Papers.

[56]  Yuyuan Tian,et al.  Measurement of the quantum capacitance of graphene. , 2009, Nature nanotechnology.

[57]  Hanan Dery,et al.  A two-dimensional spin field-effect switch , 2016, Nature Communications.

[58]  Massimiliano Di Ventra,et al.  The parallel approach , 2013 .

[59]  A. Geim,et al.  Two-dimensional gas of massless Dirac fermions in graphene , 2005, Nature.

[60]  Kaushik Roy,et al.  SPINDLE: SPINtronic Deep Learning Engine for large-scale neuromorphic computing , 2014, 2014 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED).

[61]  Yiran Chen,et al.  Memristor Crossbar-Based Neuromorphic Computing System: A Case Study , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[62]  Takashi Taniguchi,et al.  Spin Lifetimes Exceeding 12 ns in Graphene Nonlocal Spin Valve Devices. , 2016, Nano letters.

[63]  Peter R. Kinget,et al.  A programmable analog cellular neural network CMOS chip for high speed image processing , 1995, IEEE J. Solid State Circuits.

[64]  Robert Hecht-Nielsen,et al.  Theory of the backpropagation neural network , 1989, International 1989 Joint Conference on Neural Networks.